86 Professor J. W. Judd [May 2, 



that the district of which Krakatoa is the centre was likely to again 

 become the scene of volcanic disturbance, and on the 20th of May, 1883, 

 Krakatoa again burst into activity. This eruption, wHch was of 

 moderate violence, continued for about three months, resulting in the 

 formation of two cinder-cones of considerable size, and the scattering 

 of large quantities of pumice over the surface of the ocean. The 

 materials of Krakatoa and the neighbouring volcanoes are almost 

 entirely pumiceous in character, being formed through Ihe distension 

 by gases of a hypersthene-augite-andesite with an exceptionally large 

 proportion of a highly vitreous base. 



On the afternoon of August 26th, the volcano passed from the stage 

 of continued moderate activity, to a paroxysm of great violence. 

 During this paroxysm, the whole district for 100 miles around the 

 volcano was enveloped in intense darkness, produced by the enormous 

 clouds of volcanic dust thrown into the atmosphere, some of which fell 

 at distances of over 1000 miles from Krakatoa. Our knowledge of what 

 took place during this terrible outbreak is derived from the reports of 

 the few survivors in the towns on the shores of the strait, from the 

 logs of various ships, three of which were actually within the strait at 

 the time of the eruption, from the indication afforded by self-recording 

 instruments at Batavia and more distant localities, and by a comparison 

 of the condition of tbe volcano before and after the eruption. 



It appears that during the height of the eruption the detonations of 

 the volcano increased in number and violence till they blended in a 

 continuous roar, and that enormous quantities of steam with pumice and 

 dust were flung to very great heights in the atmosphere. The result 

 of this action was the blowing away, not only the two cones recently 

 formed, but of the greater part of the old volcano, leaving a vast 

 crateral hollow more than 1000 feet in depth. As to the earthquake 

 shocks accompanying this paroxysm, the accounts are very con- 

 flicting. There was unfortunately no seismograph at Batavia, but the 

 magnetograph-records seem to indicate that considerable seismic dis- 

 turbance took place during the whole time of the eruption. The 

 precisely similar instrument at Kew Observatory recorded in the same 

 w^ay the small earthquake of April 22nd, 1884. 



No doubt exists, however, as to the nature and magnitude of the 

 sea-waves produced by these great concussions. During the whole 

 time of the eruption, the ocean was thrown into a state of violent oscil- 

 lations, the oscillations increasing in height as the eruptive action 

 became more intense. In the narrow and shallow eastern throat of 

 the Sunda Strait, and in the deep gulfs of Lampong, Semangka, and 

 Welcome Bay, these vibratory movements of the water were converted 

 into waves of translation of great height and destructiveness. West- 

 ward, however, they were propagated across the Indian Ocean, like 

 the ordinary oceanic tidal wave, at rates varying from 350 to 500 

 miles per hour, recording themselves on the tide-gauges all over the 

 world. At Mauritius and Port Elizabeth, at Aden and the principal 

 Indian ports, and even as far away as Australia, New Zealand, San 



